Aarush Sah
2,601 posts
Aarush Sah
@aarush
- the last thing you see before google obliterates your startup
Having a deep think... - We’ve ben seeing a lot of demand for Kimi K2 on @GroqInc. Happy to say that it’s now available on the Groq API at 185 tokens per second, 6x faster than any other provider (AT FULL CONTEXT)
00:00- GPT-OSS, now running on Groq. 1,200 tk/s for 20B, 536 tk/s for 120B.
- I miss o3 It was a great model, and excellent as a default chatbot. GPT-5 just seems too impartial and unopinionated. o3 had opinions and was unafraid of speaking its mind
- Introducing DeepSeek-V3.1: our first step toward the agent era! 🚀 🧠 Hybrid inference: Think & Non-Think — one model, two modes ⚡️ Faster thinking: DeepSeek-V3.1-Think reaches answers in less time vs. DeepSeek-R1-0528 🛠️ Stronger agent skills: Post-training boosts tool use and
- @ anyone DM me if u need help with anything! Many amazing people have helped me and I’d like to pay it forward as best I can. Don't pitch me anything or try to get hired - just be genuine and I’ll help where I can
- We've got it running on @GroqInc. if you want the fastest (and cheapest!) inference in the world for QwQ-32B, check out Groq :)
Today, we release QwQ-32B, our new reasoning model with only 32 billion parameters that rivals cutting-edge reasoning model, e.g., DeepSeek-R1. Blog: qwenlm.github.io/blog/qwq-32b HF: huggingface.co/Qwen/QwQ-32B ModelScope: modelscope.cn/models/Qwen/Qw… Demo: huggingface.co/spaces/Qwen/Qw… Qwen Chat: - I'm hiring interns to work with me @GroqInc. You'll be doing: evals, infra, and/or post-training Term: Fall or Winter, full-time What we're looking for: - evals/post-training experience OR really solid CS/Applied AI fundamentals - thrives in high-intensity environments and
